Studies completed, one demonstrating the effect of metabolic alkalosis on oxygen consumption, and the second demonstrating a decrease in oxygen availability to the tissues when animals with large intrapulmonary shunting of blood from the right to the left sides of the circulation (atelectasis, pneumonia, congenital cardiac defects) are hyperventilated, have proven our original hypothesis that current therapeutic regimens directed against arterial hypoxemia, hypercarbia, and metabolic acidosis may jeopardize tissue oxygen utilization. In addition to the increased oxygen demand which (because of the shunt) inevitably leads to decreased arterial oxygen transport the situation is further compounded by our finding that such therapy results in a further increase in the amount of blood shunted from the ventilated alveoli. There are two major ways the shunt could increase; more of the pulmonary circulation may be directed to portions of the lung not already being ventilated (by attenuating hypoxic vasoconstriction), or this effect could occur by modifying the ventilation-perfusion ratios of the healthy portions of the lung. This distinction is crucial to designing proper therapy (including such basic issues as route of administration, i.e. via the airway, or the circulation to reach unventilated areas) and is the first objective of our continuing investigation. In addition to investigating conditions at the alveolar-blood interface our second objective is to characterize the means by which alkalosis increases oxygen consumption and to explore a strategy of controlled acidosis to counteract this adverse effect.